الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 130 |  |  |
| | | from | 130 | | |
Abstract
The thesis presents an overview of the scheduling of workflow’s tasks on the cloud environments and a new workflow scheduling algorithm called (IWDC) was developed . This algorithm is an extension for the Intelligent Water Drops algorithm (IWD), where IWD is used first to retrieve a list of the candidate execution paths within the workflow, ordered by their expected need for execution power. These paths are fed to the next phase, which is responsible for the allocation of their tasks to the available virtual machines, where the heaviest candidate paths are allocated to the fastest VMs. For evaluating its efficiency, the algorithm was developed and embedded within the WorkflowSim simulator, then; several experiments were conducted to compare it with the recent workflow scheduling algorithms. In the experiments, different types of workflows with different sizes and different architectures were used, where various cost plans were applied to schedule these workflows on several numbers of VMs with different configurations. The results proved the high ability of the IWDC algorithm, in the most of the experiments, to minimize the Makspan and reduce the execution costs in comparisons with the other competing algorithms. The thesis is organized as follows:Chapter 1: This chapter introduces the main objectives of the thesis, the research problem and the hypotheses to solve it.Chapter 2: This chapter presents an introduction to workflow management system, and reviews the recent algorithms that have been developed to schedule scientific workflow in cloud computing environments Chapter 3: In this chapter, an extended Meta-Heuristic scheduling algorithm is introduced and implemented based on the Intelligent Water DROP algorithm. This modified algorithm aims to reduce both the total execution time (makespan) and the total cost of implementing the workflow application on the cloud computing environment. The IWDC algorithm uses optimization equations to improve the scheduling process, these algorithms are derived from the behavior of the water drops’ velocity and the soil associated with them as they move. Chapter 4: In this chapter, experiments were conducted to evaluate the performance of the (IWDC) algorithm. The Workflowsim was used to simulate the cloud computing environment. We used a scientific workflow in our experiments including Montage, Epignomic, Ligo, Sipht, and Cybershake. Simulation experiments showed that IWDC has better performance than other algorithms most of the time, and it achieves low cost and makespan regardless of structure, type and volume of the workflow.Chapter 5: This chapter includes the conclusions derived from this thesis and some ideas that can be applied in the future to enhance the performance of the IWDC algorithm presented and developed in this thesis.